Genome-wide profiling of DNA methylation at single-base resolution based on MeDIP-bisulfite high-throughput sequencing and ridge regression (RNA)

Unraveling complexity of DNA methylome is essential to decipher DNA methylation mechanism in life. However, this has been subjected to technological constraints to balance between cost and accurate measurement of the DNA methylation level. In this study, by innovatively introducing C-hydroxylmethylated adapters, we have developed MeDIP-Bisulfite sequencing (MB-seq), which could obtain DNA methylome of repertoire CpGs at single-base resolution. We found MB-seq only costs 10% of MethylC-seq, but covers 85% of total CpGs in human genome. Unlike absolute methylation levels determined by MethylC-seq and RRBS, MB-seq presented relative methylation levels that are linearly inflated. This has enlightened us to develop a MB-seq corresponding correction method for methylation level based on ridge regression, which integrates the data of MB-seq and RRBS to predict the methylation level of total 28.2 million CpGs on human genome with high accuracy (Pearson correlation coefficient, PCC=0.90). Moreover, by employing MB-seq, we generated the DNA methylome of an ovarian epithelial cell line (T29) and its oncogenic counterpart (T29H), respectively. After ridge regression, we identified 131,790 differential methylation regions (DMRs) with high accuracy between T29 and T29H, far more than 7,567 obtained from RRBS. Taken together, our result demonstrated that the MB-seq combined with ridge regression is a wide applicable approach for profiling of DNA methylome. Overall design: Total RNAs were extracted from T29 and T29H with RNeasy Mini Kit (QIAGEN, Germany). RNA quality was quality-controlled by Bioanalyser 2100 (RNA nano kits, Agilent). mRNA-Seq libraries were generated from total RNA with polyA+ selection of mRNA using the TruSeq RNA Sample Prep Kit v2 (Illumina, San Diego, CA), and then subjected to transcriptome sequencing on the Illumina Hiseq 2000

2

Cai W, Mao F, Teng H, Cai T, Zhao F, Wu J, Sun ZS